RFR: 8363620: AArch64: reimplement emit_static_call_stub()

[Andrew Haley]

On Tue, 5 Aug 2025 21:40:27 GMT, Dean Long <dlong at openjdk.org> wrote:

> In the past we would have scheduled the load for x8 earlier to avoid a latency hit for using the value in the next instruction:
> 
> ```
> ldr x8, Label_entry
> ldr x12, Label_data
> br x8
> ```
> 
> Is this still helpful on modern aarch64 hardware?

Maybe on something very small, such as the Cortex-A34. It doesn't hurt larger out-of-order cores, so I agree with you that we should reorder this.

-------------

PR Comment: https://git.openjdk.org/jdk/pull/26638#issuecomment-3158299392